diff options
author | Michael Drake <tlsa@netsurf-browser.org> | 2021-05-16 14:35:30 +0100 |
---|---|---|
committer | Michael Drake <michael.drake@codethink.co.uk> | 2021-05-16 14:35:30 +0100 |
commit | d101b2bb6dc98050f8f1b04d9d2bfeeff5a120c7 (patch) | |
tree | e62cf6cacc4b33a40672b8e1696817a16144c997 | |
parent | 269b0047aa506be93637e93d109a34f4bd7cefc4 (diff) | |
download | libparserutils-d101b2bb6dc98050f8f1b04d9d2bfeeff5a120c7.tar.gz libparserutils-d101b2bb6dc98050f8f1b04d9d2bfeeff5a120c7.tar.bz2 |
Buffer: Optimise to minimise memmove shuffles.
Previously the data in the linear buffer was always stored at
the start of the allocation, pointed to by `buffer->data`.
This was achieved by memmoving every time data was consumed
from the front.
Now the allocation is pointed to by `buffer->alloc`, and the
start of the data is pointed to by `buffer->data` (as before).
This means client code does not need to change to get at the
data.
The advantage comes when we discard the from the start of the
buffer, when some data is consumed. We now simply advance the
data pointer by the number of bytes to be discarded, and reduce
the buffer length by the same amount.
If the used portion of the buffer now fits between the start
of the allocation and the current start of the data, it is
memcpyed to the allocation start, otherwise it is left alone.
This is a significant optimisation when the size of the chunk
is large, such as when loading from disc. (When the first
(only) "chunk" is just the whole file.
-rw-r--r-- | include/parserutils/utils/buffer.h | 1 | ||||
-rw-r--r-- | src/utils/buffer.c | 86 |
2 files changed, 71 insertions, 16 deletions
diff --git a/include/parserutils/utils/buffer.h b/include/parserutils/utils/buffer.h index 9425070..5f8f696 100644 --- a/include/parserutils/utils/buffer.h +++ b/include/parserutils/utils/buffer.h @@ -18,6 +18,7 @@ extern "C" struct parserutils_buffer { + uint8_t *alloc; uint8_t *data; size_t length; size_t allocated; diff --git a/src/utils/buffer.c b/src/utils/buffer.c index 716e67c..5e0f58c 100644 --- a/src/utils/buffer.c +++ b/src/utils/buffer.c @@ -30,12 +30,13 @@ parserutils_error parserutils_buffer_create(parserutils_buffer **buffer) if (b == NULL) return PARSERUTILS_NOMEM; - b->data = malloc(DEFAULT_SIZE); - if (b->data == NULL) { + b->alloc = malloc(DEFAULT_SIZE); + if (b->alloc == NULL) { free(b); return PARSERUTILS_NOMEM; } + b->data = b->alloc; b->length = 0; b->allocated = DEFAULT_SIZE; @@ -55,13 +56,59 @@ parserutils_error parserutils_buffer_destroy(parserutils_buffer *buffer) if (buffer == NULL) return PARSERUTILS_BADPARM; - free(buffer->data); + free(buffer->alloc); free(buffer); return PARSERUTILS_OK; } /** + * Get current data offset within buffer's allocation. + * + * \param[in] buffer The buffer object. + * \return data offset in bytes. + */ +static inline size_t get_offset(parserutils_buffer *buffer) +{ + return buffer->data - buffer->alloc; +} + +/** + * Try moving the data to the start of the allocation. + * + * \param[in] buffer The buffer object. + */ +static inline void try_rebase(parserutils_buffer *buffer) +{ + if (get_offset(buffer) >= buffer->length) { + memcpy(buffer->alloc, buffer->data, buffer->length); + buffer->data = buffer->alloc; + } +} + +/** + * Ensure that the buffer has enough space at the end to add len bytes. + * + * \param[in] buffer The buffer object. + * \param[in] len Number of bytes to ensure there is space for. + * \return PARSERUTILS_OK on success, appropriate error otherwise. + */ +static inline parserutils_error ensure_space( + parserutils_buffer *buffer, + size_t len) +{ + try_rebase(buffer); + + while (len >= buffer->allocated - buffer->length - get_offset(buffer)) { + parserutils_error error = parserutils_buffer_grow(buffer); + if (error != PARSERUTILS_OK) + return error; + } + + return PARSERUTILS_OK; +} + +/** * Append data to a memory buffer * * \param buffer The buffer to append to @@ -72,11 +119,9 @@ parserutils_error parserutils_buffer_destroy(parserutils_buffer *buffer) parserutils_error parserutils_buffer_append(parserutils_buffer *buffer, const uint8_t *data, size_t len) { - while (len >= buffer->allocated - buffer->length) { - parserutils_error error = parserutils_buffer_grow(buffer); - if (error != PARSERUTILS_OK) - return error; - } + parserutils_error error = ensure_space(buffer, len); + if (error != PARSERUTILS_OK) + return error; memcpy(buffer->data + buffer->length, data, len); @@ -97,17 +142,17 @@ parserutils_error parserutils_buffer_append(parserutils_buffer *buffer, parserutils_error parserutils_buffer_insert(parserutils_buffer *buffer, size_t offset, const uint8_t *data, size_t len) { + parserutils_error error; + if (offset > buffer->length) return PARSERUTILS_BADPARM; if (offset == buffer->length) return parserutils_buffer_append(buffer, data, len); - while (len >= buffer->allocated - buffer->length) { - parserutils_error error = parserutils_buffer_grow(buffer); - if (error != PARSERUTILS_OK) - return error; - } + error = ensure_space(buffer, len); + if (error != PARSERUTILS_OK) + return error; memmove(buffer->data + offset + len, buffer->data + offset, buffer->length - offset); @@ -133,6 +178,13 @@ parserutils_error parserutils_buffer_discard(parserutils_buffer *buffer, if (offset >= buffer->length || offset + len > buffer->length) return PARSERUTILS_BADPARM; + if (offset == 0) { + buffer->data += len; + buffer->length -= len; + try_rebase(buffer); + return PARSERUTILS_OK; + } + memmove(buffer->data + offset, buffer->data + offset + len, buffer->length - (len + offset)); @@ -149,11 +201,13 @@ parserutils_error parserutils_buffer_discard(parserutils_buffer *buffer, */ parserutils_error parserutils_buffer_grow(parserutils_buffer *buffer) { - uint8_t *temp = realloc(buffer->data, buffer->allocated * 2); + size_t offset = get_offset(buffer); + uint8_t *temp = realloc(buffer->alloc, buffer->allocated * 2); if (temp == NULL) return PARSERUTILS_NOMEM; - buffer->data = temp; + buffer->alloc = temp; + buffer->data = buffer->alloc + offset; buffer->allocated *= 2; return PARSERUTILS_OK; @@ -180,7 +234,7 @@ parserutils_error parserutils_buffer_randomise(parserutils_buffer *buffer) /* Leak the buffer's current data, so we don't reuse it */ /* buffer->alloc(buffer->data, 0, buffer->pw); */ - buffer->data = temp; + buffer->alloc = buffer->data = temp; #endif |